Nonlinear Effort-Time Dynamics of Student Engagement in a Web-Based Learning Platform: A Person-Oriented Transition Analysis
Saved in:
| Title: | Nonlinear Effort-Time Dynamics of Student Engagement in a Web-Based Learning Platform: A Person-Oriented Transition Analysis |
|---|---|
| Language: | English |
| Authors: | Elissavet Papageorgiou (ORCID |
| Source: | Journal of Learning Analytics. 2025 12(2):237-258. |
| Availability: | Society for Learning Analytics Research. 121 Pointe Marsan, Beaumont, AB T4X 0A2, Canada. Tel: +61-429-920-838; e-mail: info@solaresearch.org; Web site: https://learning-analytics.info/index.php/JLA/index |
| Peer Reviewed: | Y |
| Page Count: | 53 |
| Publication Date: | 2025 |
| Document Type: | Journal Articles Reports - Research |
| Education Level: | Higher Education Postsecondary Education |
| Descriptors: | Learner Engagement, Student Behavior, Electronic Learning, Web Based Instruction, Foreign Countries, College Students, Learning Analytics, Behavior Patterns, Time Factors (Learning), Behavior Change |
| Geographic Terms: | Netherlands |
| ISSN: | 1929-7750 |
| Abstract: | Behavioural engagement as a predictor of academic success hinges on the interplay between effort and time. Exploring the longitudinal development of engagement is vital for understanding adaptations in learning behaviour and informing educational interventions. However, person-oriented longitudinal studies on student engagement are scarce. Moreover, online engagement metrics are rarely grounded in theory and often result in simplified descriptions overlooking the complexity of engagement processes. This study applies a theory-based operationalization of behavioural engagement to examine the log data of 236 students in a web-based learning platform. We explored (1) whether weekly profiles based on distinct engagement patterns can be identified and (2) how students transition across profiles over time. Hierarchical clustering yielded one Inactive and six active profiles (Fast-Learners, Regular-Learners, Average-Engagement, Minimalists, Struggling-Learners, and Procrastinators). Results suggest heterogeneity in profile emergence, with effective engagement characterized by alignment with the course deadlines. Process mining revealed changes in profile membership across weeks. Profile transitions revealed relative stability among effective groups and greater fluctuation among low-time profiles. By investigating the complexity and temporality of engagement in online learning, our findings provide insights for developing personalized learning support through training artificial intelligence applications and informing learning analytics dashboards. |
| Abstractor: | As Provided |
| Entry Date: | 2025 |
| Accession Number: | EJ1483354 |
| Database: | ERIC |
| Abstract: | Behavioural engagement as a predictor of academic success hinges on the interplay between effort and time. Exploring the longitudinal development of engagement is vital for understanding adaptations in learning behaviour and informing educational interventions. However, person-oriented longitudinal studies on student engagement are scarce. Moreover, online engagement metrics are rarely grounded in theory and often result in simplified descriptions overlooking the complexity of engagement processes. This study applies a theory-based operationalization of behavioural engagement to examine the log data of 236 students in a web-based learning platform. We explored (1) whether weekly profiles based on distinct engagement patterns can be identified and (2) how students transition across profiles over time. Hierarchical clustering yielded one Inactive and six active profiles (Fast-Learners, Regular-Learners, Average-Engagement, Minimalists, Struggling-Learners, and Procrastinators). Results suggest heterogeneity in profile emergence, with effective engagement characterized by alignment with the course deadlines. Process mining revealed changes in profile membership across weeks. Profile transitions revealed relative stability among effective groups and greater fluctuation among low-time profiles. By investigating the complexity and temporality of engagement in online learning, our findings provide insights for developing personalized learning support through training artificial intelligence applications and informing learning analytics dashboards. |
|---|---|
| ISSN: | 1929-7750 |
